DESCRIPTION (Applicant's Abstract): The overall goal of this project is to develop a sensitive, precisely defined animal model of impulsive behavior. The model is based on the assumption that impulsive individuals are relatively insensitive to delayed and/or uncertain consequences (rewards and punishers). Devaluation of delayed and/or uncertain consequences causes impulsive individuals to be relatively more influenced by immediate certain consequences. The model measures sensitivity to consequences in terms of "discounted" value such that impulsive individuals discount the value of delayed and/or uncertain rewards and punishers more then nonimpulsive individuals. In the animal model the value of delayed and/or uncertain consequences is assessed quantitatively in rats. Eight experiments are proposed to asses the reliability and validity of the discounting procedure. In order to validate the model, manipulations which cause impulsive behaviors in humans will be tested. Thus, the effects of serotonin (5HT) lesions; and administration of amphetamine, ethanol and diazepam will be tested in all 8 experiments. Experiments 1 & 2 are control experiments which will assess the reliability of the procedure with different amounts of reward and punishment. It is predicted that 5HT lesions and drugs will not affect discounting in Experiments 1 & 2 because these experiments do not involve delay or uncertainty. Experiments 3 & 4 examine discounting of the value of rewarding and punishing consequences by delay. Experiments 5 & 6 examine discounting of rewarding and punishing consequences by uncertainty. Experiments 7 & 8 examine discounting of rewarding and punishing consequences by the combination of delay and uncertainty. It is predicted that 5HT lesions and amphetamine, ethanol, and diazepam will increase discounting in experiments 3, 4, 5, 6, 7, & 8. Establishing an animal model of impulsive human behavior will enable researchers to more systematically explore the neurochemical and pharmacological basis of impulsive behavior.